Induction of a T lymphocyte response is a critical initial step in a host's immune response. Activation of T cells results in T cell proliferation, cytokine production by T cells and generation of T cell-mediated effector functions. T cell activation requires an antigen-specific signal, often called a primary activation signal, which results from stimulation of a clonally-distributed T cell receptor (hereafter TcR) present on the surface of the T cell. This antigen-specific signal is usually in the form of an antigenic peptide bound either to a major histocompatibility complex (hereafter MHC) class I protein or an MHC class II protein present on the surface of an antigen presenting cell (hereafter APC). CD4+ T cells recognize peptides associated with class II molecules. Class II molecules are found on a limited number of cell types, primarily B cells, monocytes/macrophages and dendritic cells, and, in most cases, present peptides derived from proteins taken up from the extracellular environment. In contrast, CD8+ T cells recognize peptides associated with class I molecules. Class I molecules are found on almost all cell types and, in most cases, present peptides derived from endogenously synthesized proteins. For a review see Germain, R., Nature 322, 687-691 (1986).
It has now been established that, in addition to an antigen-specific primary activation signal, T cells also require a second, non-antigen specific, signal to induce full T cell proliferation and/or cytokine production. This phenomenon has been termed costimulation. Mueller, D. L., et al., Annu. Rev. Immunol. 7, 445-480 (1989). Like the antigen-specific signal, the costimulatory signal is triggered by a molecule on the surface of the antigen presenting cell. A costimulatory molecule, the B lymphocyte antigen B7, has been identified on activated B cells and other APCs. Freeman, G. J., et al., J Immunol. 139, 3260-3267 (1987); Freeman, G. J., et al., J Immunol. 143, 2714-2722 (1989). Binding of B7 to a ligand on the surface of T cells provides costimulation to the T cell. Two structurally similar T cell-surface receptors for B7 have been identified, CD28 and CTLA-4. Aruffo, A. and Seed, B., Proc. Natl. Acad. Sci. USA 84, 8573-8577 (1987); Linsley, P. S., et al., J Exp. Med 173, 721-730, (1991); Brunet, J. F., et al., Nature 328, 267-270 (1987); Brunet, J. F., et al., Immunol Rev. 103, 21-36 (1988). CD28 is expressed constitutively on T cells and its expression is upregulated by activation of the T cell, such as by interaction of the TcR with an antigen-MHC complex. In contrast, CTLA4 is undetectable on resting T cells and its expression is induced by activation.
A series of experiments have shown a functional role for a T cell activation pathway stimulated through the CD28 receptor. Studies using blocking antibodies to B7 and CD28 have demonstrated that these antibodies can inhibit T cell activation, thereby demonstrating the need for stimulation via this pathway for T cell activation. Furthermore, suboptimal polyclonal stimulation of T cells by phorbol ester or anti-CD3 antibodies can be potentiated by crosslinking of CD28 with anti-CD28 antibodies. Engagement of the TcR by an MHC molecule/peptide complex in the absence of the costimulatory B7 signal can lead to T cell anergy rather than activation. Damle, N. K., et al., Proc. Natl. Acad. Sci. USA 78, 5096-5100 (1981); Lesslauer, W., et al., Eur. J. Immunol. 16, 1289-1295 (1986); Gimmi, C. D., et al., Proc. Natl. Acad. Sci. USA 88, 6575-6579 (1991); Linsley, P. S., et al., J Exp. Med. 173, 721-730 (1991); Koulova, L., et al., J Exp. Med. 173, 759-762 (1991); Razi-Wolf, Z., et al., Proc. Natl. Acad Sci. USA 89, 4210-4214 (1992).
Malignant transformation of a cell is commonly associated with phenotypic changes in the cell. Such changes can include loss or gain of expression of some proteins or alterations in the level of expression of certain proteins. It has been hypothesized that in some situations the immune system may be capable of recognizing a tumor as foreign and, as such, could mount an immune response against the tumor. Kripke, M., Adv. Cancer Res. 34, 69-75 (1981). This hypothesis is based in part on the existence of phenotypic differences between a tumor cell and a normal cell, which is supported by the identification of tumor associated antigens (hereafter TAAs). Schreiber, H., et al. Ann. Rev. Immunol. 6, 465-483 (1988). TAAs are thought to distinguish a transformed cell from its normal counterpart. Three genes encoding TAAs expressed in melanoma cells, MAGE-1, MAGE-2 and MAGE-3, have recently been cloned. van der Bruggen, P., et al. Science 254, 1643-1647 (1991). That tumor cells under certain circumstances can be recognized as foreign is also supported by the existence of T cells which can recognize and respond to tumor associated antigens presented by MHC molecules. Such TAA-specific T lymphocytes have been demonstrated to be present in the immune repertoire and are capable of recognizing and stimulating an immune response against tumor cells when properly stimulated in vitro. Rosenberg, S. A., et al. Science 233, 1318-1321 (1986); Rosenberg, S. A. and Lotze, M. T. Ann. Rev. Immunol. 4, 681-709 (1986).
However, in practice, tumors in vivo have generally not been found to be very immunogenic and appear to be capable of evading immune response. This may result from an inability of tumor cells to induce T cell-mediated immune responses. Ostrand-Rosenberg, S., et al., J Immunol. 144, 4068-4071 (1990); Fearon, E. R., et al., Cell 60, 397-403 (1990). A method for increasing the immunogenicity of a tumor cell in vivo would be therapeutically beneficial.